检索结果-内蒙古大学图书馆

IEEE 26th international conference on Intelligent Transportation Systems (ITSC)

作者： Wang, Ruiqi Ji, Ang Li, Tao Sun, Zhanbo Fu, Zhijian Southwest Jiaotong Univ Sch Transportat & Logist Natl Engn Lab Integrated Transportat Big Data Chengdu Sichuan Peoples R China

ISBN: (纸本)9798350399462

The paper introduces an interpretable machine learning technique SHAP (SHapley Additive exPlanation) to analyze the vehicle yielding behaviors during pedestrian-vehicle interactions at unsignalized intersections. The study first extracts trajectory data from drone videos and then exploits machine learning methods to construct the yielding classification model. The results indicate that Random Forest (RF) outperforms Support Vector machine (SVM), Gradient Boosting machine (GBM), and eXtreme Gradient Boosting (XGBoost), achieving the best classification performance with an area under the ROC curve (AUC) of 0.934. Finally, the SHAP algorithm is fused with RF to improve the model interpretability. The analysis reveals that the distances between vehicles and pedestrians make the most significant impact on vehicle yielding behavior. Furthermore, it is found that traffic-related variables exhibit non-linear and threshold effects on vehicle yielding.

关键词： Vehicles

来源：评论

学校读者我要写书评

暂无评论

A Model-Agnostic Approach for learning with Noisy Labels of Arbitrary Distributions 38

A Model-Agnostic Approach for Learning with Noisy Labels of ...

引用

38th IEEE international conference on data engineering (ICDE)

作者： Hao, Shuang Li, Peng Wu, Renzhi Chu, Xu Beijing Jiaotong Univ Sch Comp & Informat Technol Beijing Peoples R China Beijing Key Lab Traff Data Anal & Min Beijing Peoples R China Georgia Inst Technol Atlanta GA 30332 USA

ISBN: (数字)9781665408837

ISBN: (纸本)9781665408837

Most real-world datasets contain label noise, which can negatively affect downstream ML models trained on them. To deal with this problem, one can clean the mislabeled data before training, which is not only time-consuming and expensive but also requires domain expertise. Another approach is to use a noiserobust ML training algorithm. However, existing methods have some prerequisites that may not be practical in many applications (e.g., they are tied to specific downstream model architecture or they are applicable to specific noise distributions). In this paper, we propose a model-agnostic approach for learning with noisy labels of arbitrary distributions. In particular, our approach can work with any gradient descent optimization based machine learning model and deal with any label noise distribution. We achieve them by proposing two theoretically grounded noise-robust loss functions (for different noise distributions), and we are able to automatically decide which loss function to use based on a novel noise setting detection module. We directly learn the required hyper-parameters in the loss functions via meta-learning technique to minimize the loss on a given small clean validation set, and propose several strategies to improve the efficiency of training. Experiments on multiple datasets with both real-world and injected label noise show that our method performs better than state-of-the-art approaches.

关键词： Training conferences machine learning data engineering Noise robustness Noise measurement Optimization

来源：评论

学校读者我要写书评

暂无评论

data AUGMENTATION OF engineering DRAWINGS FOR data-DRIVEN COMPONENT SEGMENTATION

DATA AUGMENTATION OF ENGINEERING DRAWINGS FOR DATA-DRIVEN CO...

引用

ASME international Design engineering Technical conferences / 42nd Annual Computers and Information in engineering conference (IDETC-CIE) / 48th Design Automation conference (DAC)

作者： Zhang, Wentai Chen, Quan Koz, Can Xie, Liuyue Regmi, Amit Yamakawa, Soji Furuhata, Tomotake Shimada, Kenji Kara, Levent Burak Carnegie Mellon Univ Pittsburgh PA 15213 USA

ISBN: (纸本)9780791886229

We present a new data generation method to facilitate an automatic machine interpretation of 2D engineering part drawings. While such drawings are a common medium for clients to encode design and manufacturing requirements, a lack of computer support to automatically interpret these drawings necessitates part manufacturers to resort to laborious manual approaches for interpretation which, in turn, severely limits processing capacity. Although recent advances in trainable computer vision methods may enable automatic machine interpretation, it remains challenging to apply such methods to engineering drawings due to a lack of labeled training data. As one step toward this challenge, we propose a constrained data synthesis method to generate an arbitrarily large set of synthetic training drawings using only a handful of labeled examples. Our method is based on the randomization of the dimension sets subject to two major constraints to ensure the validity of the synthetic drawings. The effectiveness of our method is demonstrated in the context of a binary component segmentation task with a proposed list of descriptors. An evaluation of several image segmentation methods trained on our synthetic dataset shows that our approach to new data generation can boost the segmentation accuracy and the generalizability of the machine learning models to unseen drawings.

关键词： engineering drawings Image segmentation Computers machinery Dimensions Manufacturing Design machine learning

来源：评论

学校读者我要写书评

暂无评论

DNN-based Secure Remote Patient data Analysis Framework for Improving Human Life Expectancy in Healthcare 4.0 2

DNN-based Secure Remote Patient Data Analysis Framework for ...

引用

2nd international conference on machine learning and data engineering, icmlde 2023

作者： Darji, Krisha Ramoliya, Fenil Kakkar, Riya Gupta, Rajesh Tanwar, Sudeep Garg, Deepak Department of Computer Science and Engineering Institute of Technology Nirma University Ahmedabad India School of Computer Science and Artificial Intelligence SR University Telangana Warangal506371 India

The evolution of healthcare from its early beginnings to the recent healthcare 4.0 revolution has remarkably improved human life and living standards. The integration of emerging technologies has played a pivotal role in this progress. Notably, remote analysis of patient data has emerged as a promising approach, enabling telemedicine and remote patient monitoring enhancing healthcare accessibility and efficiency. However, the vulnerabilities of patient data to malicious network attacks pose critical challenges in the healthcare industry, potentially compromising patients' safety and undermining trust in the system. Thus, we have introduced a cutting-edge deep neural network (DNN) model to mitigate the risks associated with remote patient data transfer in healthcare 4.0 by bifurcating the data into malicious or non-malicious. The primary objective of the proposed framework is to ensure the secure and private communication of patient data, thereby fostering a more dependable and trustworthy healthcare ecosystem. Finally, the proposed framework is evaluated against various standard metrics such as accuracy, loss considering binary cross-entropy, receiver operating characteristic (ROC) curve, precision-recall curve and confusion matrix. The implications of the proposed framework offer data security to the healthcare system as it contributes to creating a more resilient and dependable ecosystem, thus promoting better patient care and ultimately elevating the overall life expectancy and well-being of individuals. © 2024 Elsevier B.V.. All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Enhancing Intracranial Hemorrhage Diagnosis through Deep learning Models 2

Enhancing Intracranial Hemorrhage Diagnosis through Deep Lea...

引用

2nd international conference on machine learning and data engineering, icmlde 2023

作者： Malik, Payal Dureja, Ajay Dureja, Aman Rathore, Rajkumar Singh Malhotra, Nisha New Delhi India Rohini New Delhi India Department of Computer Science Cardiff School of Technologies United Kingdom

Timely diagnosis is crucial for the successful treatment of a serious medical condition like brain hemorrhage. Deep learning algorithms have shown great promise in applications for medical image analysis, like the identification of brain hemorrhages. The goal of this study is to assess how well various deep learning algorithms can detect brain hemorrhages. Using a suitable dataset, the study evaluates the computational efficiency, accuracy, sensitivity, and specificity of the selected algorithms. The results demonstrate the potential of deep learning models to assist physicians in identifying this potentially fatal condition and demonstrate how well they can identify brain hemorrhages. The study's findings improve automated brain hemorrhage detection technology, improving patient outcomes and the efficiency of healthcare delivery. EfficientNetB3 typically achieves higher accuracy due to its increased model complexity. Despite its increased complexity, EfficientNetB3 is still more parameter-efficient and computationally efficient than many alternative architectures. EfficientNetB3's strong performance and feature extraction capabilities make it a good choice for transfer learning tasks. Out of all the models implemented in this paper, the proposed model with EfficientNetB3 gave the best accuracy for training as well as validation i.e. 99.95% and 93.29% respectively followed by EfficientNetB2, ResNet, SEResNext and ResNext. © 2024 Elsevier B.V.. All rights reserved.

关键词： Medical imaging

来源：评论

学校读者我要写书评

暂无评论

An Improvised Location Aided Routing Protocol for Vehicular Ad-hoc Network using Reinforced learning

An Improvised Location Aided Routing Protocol for Vehicular ...

引用

2022 international conference on machine learning, Big data, Cloud and Parallel Computing, COM-IT-CON 2022

作者： Kumar, Arvind Dixit, Prashant Tyagi, S.S. Manav Rachna International Institute of Research & Studies Computer Science & Engineering Faridabad India IIMT College of Engineering Greater Noida India

ISBN: (数字)9781665496025

ISBN: (纸本)9781665496025

Vehicular Ad-hoc Networks (VANETs) are dynamic networks formed among the vehicles travelling on road. They are prone to change in network topologies very quickly due to limited range of randomly moving vehicles. A vehicle is in range at a time and out of range the other moment. This limitation paved the way for development of routing protocols for them which are different from the routing protocols for mobile ad hoc networks. Since most of the vehicles quite often travel within a city, the routing protocol VANETs can also be made intelligent with the use of deep reinforcement learning (RL). With the help of deep reinforcement learning, the routing protocol will keep on learning with their experience and take suitable routing decision in the future. © 2022 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

FedBR: Improving Federated learning on Heterogeneous data via Local learning Bias Reduction 40

FedBR: Improving Federated Learning on Heterogeneous Data vi...

引用

40th international conference on machine learning

作者： Guo, Yongxin Tang, Xiaoying Lin, Tao Chinese Univ Hong Kong Sch Sci & Engn Shenzhen 518172 Guangdong Peoples R China Shenzhen Inst Artificial Intelligence & Robot Soc Shenzhen Peoples R China Chinese Univ Hong Kong Shenzhen Guangdong Prov Key Lab Future Networks Intelligen Shenzhen 518172 Peoples R China Westlake Univ Res Ctr Ind Future Hangzhou Peoples R China Westlake Univ Sch Engn Hangzhou Peoples R China

Federated learning (FL) is a way for machines to learn from data that is kept locally, in order to protect the privacy of clients. This is typically done using local SGD, which helps to improve communication efficiency. However, such a scheme is currently constrained by slow and unstable convergence due to the variety of data on different clients' devices. In this work, we identify three under-explored phenomena of biased local learning that may explain these challenges caused by local updates in supervised FL. As a remedy, we propose FedBR, a novel unified algorithm that reduces the local learning bias on features and classifiers to tackle these challenges. FedBR has two components. The first component helps to reduce bias in local classifiers by balancing the output of the models. The second component helps to learn local features that are similar to global features, but different from those learned from other data sources. We conducted several experiments to test FedBR and found that it consistently outperforms other SOTA FL methods. Both of its components also individually show performance gains. Our code is available at https: //***/lins-lab/fedbr.

关键词： Balancing

来源：评论

学校读者我要写书评

暂无评论

Research on construction of learner portrait based on learning analytics technology

Research on construction of learner portrait based on learni...

引用

2022 international conference on Artificial Intelligence and Industrial Design, AIID 2022

作者： Zhou, Yan Fan, Haodong School of Mechatronics Engineering Harbin Institute of Technology Harbin150001 China

ISBN: (纸本)9781510663893

With the development of big data, artificial intelligence, and wearable technology, the data generated by learners in the learning process can be fully recorded and stored. Using these data to study the characteristics of different types of learners and provide them with precise and personalized learning guidance is an urgent problem to be solved. This paper proposes a learner portrait construction method based on learning analysis technology. Firstly, the case frame relationship with "learning" as the central verb is established by using case grammar. Secondly, constructing a "learning-learner" feature layer with 6 sub-layers through qualitative analysis, and organizing the primary data of learners into 6 kinds of integrated data with specific meanings. Then, two methods of classifying learner types, cluster analysis and standard establishment, are proposed. Finally, giving an example of learner portrait construction. © 2023 SPIE.

关键词： data mining

来源：评论

学校读者我要写书评

暂无评论

Enhancing elderly care services through integrated sentiment analysis and knowledge reasoning: A deep learning approach

引用

international Journal of Cognitive Computing in engineering 2025年 6卷 477-494页

作者： Ai, Yongguan Chu, Shiwei Wang, Juan Xu, Nianfang School of Public Administration Anhui Vocational and Technical College Anhui Hefei230011 China School of Electronic Information Engineering Anhui University Hefei230039 China School of Computer and Information Technology Anhui Vocational and Technical College Anhui Hefei230011 China

This study proposes a pioneering integrated care model for elderly care service robots that integrates sentiment analysis and knowledge reasoning through a deep learning framework. The primary objective of this research is to address the limitations of current elderly care robots in providing emotionally intelligent and personalized care. The model utilizes advanced deep learning techniques, such as Long Short-Term Memory (LSTM) networks and Convolutional Neural Networks (CNNs), to analyze multimodal data comprising speech, facial expressions and body language. This enables the model to provide a comprehensive understanding of an elderly individual's emotional and health status. The efficacy of the model is demonstrated by its ability to enhance the precision of care decisions, improve the quality of care, user satisfaction, and system reliability. The experimental results demonstrate substantial improvements in sentiment recognition accuracy (96.5 %), reasoning accuracy (93.7 %), decision execution time (3.2 s), user satisfaction (4.9 points), and system stability (98.4 %), highlighting the transformative potential of the model in revolutionizing elderly care services. © 2025

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

machine learning in DeFi: Credit Risk Assessment and Liquidation Prediction 6

Machine Learning in DeFi: Credit Risk Assessment and Liquida...

引用

6th IEEE international conference on Blockchain and Cryptocurrency, ICBC 2024

作者： Palaiokrassas, Georgios Scherrers, Sandro Makri, Eftychia Tassiulas, Leandros Yale University Department of Electrical Engineering New HavenCT United States Yale University Yale Institute for Network Science New HavenCT United States Yale University School of Management New HavenCT United States

ISBN: (纸本)9798350316742

This paper investigates the application of machine learning for credit risk assessment in Multichain Decentralized Finance (DeFi). With DeFi expanding its scope, the need for effective credit risk evaluation becomes paramount. Our study utilizes a diverse dataset gathered from multiple blockchains, including Ethereum, and employs rigorous data preprocessing techniques. DeFi-specific features are extracted, capturing transaction-related statistics. machine learning models, such as Logistic Regression, Random Forest, XGBoost, CatBoost, LightGBM and a CNN, are deployed to predict wallet liquidations. Evaluation metrics, including accuracy, ROC curve and Area Under the Curve, demonstrate the efficacy of DeFi-related features in credit risk assessment. Furthermore, we analyze feature importance and inter-feature correlations, providing insights into critical risk factors within the DeFi ecosystem. This research contributes valuable insights to the DeFi landscape, offering data-driven approaches to credit risk management and investment strategies. Our findings hold significance for DeFi stakeholders seeking to navigate the evolving financial frontier while mitigating credit risk effectively. © 2024 IEEE.

关键词： Decentralized finance

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：